Advanced Lane Finding Project¶
The goals / steps of this project are the following:
1 - Compute the camera calibration matrix and distortion coefficients given a set of chessboard images.
2 - Apply a distortion correction to raw images.
3 - Use color transforms, gradients, etc., to create a thresholded binary image.
4 - Apply a perspective transform to rectify binary image ("birds-eye view").
5 - Detect lane pixels and fit to find the lane boundary.
6 - Determine the curvature of the lane and vehicle position with respect to center.
7 - Warp the detected lane boundaries back onto the original image.
8 - Output visual display of the lane boundaries and numerical estimation of lane curvature and vehicle position.
Libraries used in the project¶
In [1]:
import numpy as np
import cv2
import glob
import matplotlib.pyplot as plt
import matplotlib.image as mpimg
import pickle
import os
from moviepy.editor import VideoFileClip
%matplotlib inline
Helper Functions¶
In [2]:
def Calculate_Sobel(warpedimage, orient, thresh_min=0, thresh_max=255, kernelSize=3):
# 1) Convert to gray scale
gray = cv2.cvtColor(warpedimage, cv2.COLOR_RGB2GRAY)
# 2) Take the gradient in x and y separately
sobelx = cv2.Sobel(gray, cv2.CV_64F, 1, 0, ksize=kernelSize)
sobely = cv2.Sobel(gray, cv2.CV_64F, 0, 1, ksize=kernelSize)
# 3) Take the absolute value of the x and y gradients
abs_sobelx = np.absolute(sobelx)
abs_sobely = np.absolute(sobely)
abs_sobelxy = np.sqrt(sobelx**2 + sobely**2)
# 4) Get gradient as per orientation passed
if orient=='x':
grad = abs_sobelx
elif orient == 'y':
grad = abs_sobely
elif orient == 'xy':
grad= abs_sobelxy
elif orient == 'dir':
grad = np.arctan2(abs_sobely, abs_sobelx)
# 5) Scale to 8-bit (0 - 255) then convert to type = np.uint8
scaled_sobel = np.uint8(255*grad/np.max(grad))
# 6) Create a mask of 1's where the scaled gradient magnitude
# is > thresh_min and < thresh_max
binary_output = np.zeros_like(scaled_sobel)
binary_output[(scaled_sobel >= thresh_min) & (scaled_sobel <= thresh_max)] = 1
# 7) Return this mask as your binary_output image
return binary_output
In [3]:
def hls_threshold(img, channel, thresh=(0, 255)):
# Convert from RGB to HLS
hls = cv2.cvtColor(img, cv2.COLOR_RGB2HLS)
# Extract channels needed H/L/S from the image
if channel == 'h':
channelImg = hls[:,:,0]
elif channel == 'l':
channelImg = hls[:,:,1]
elif channel == 's':
channelImg = hls[:,:,2]
hlsBinary = np.zeros_like(channelImg)
hlsBinary[(channelImg > thresh[0]) & (channelImg <= thresh[1])] = 1
return hlsBinary
In [4]:
def region_of_interest(img):
mask = np.zeros_like(img)
imshape = img.shape
if len(imshape) > 2:
channel_count = img.shape[2]
ignore_mask_color = (255,) * channel_count
else:
ignore_mask_color = 255
# Define the vertices of region of interest
vertices = np.array([[(0,imshape[0]),(imshape[1]*0.48, imshape[0]*0.58), (imshape[1]*0.52, imshape[0]*0.58), (imshape[1],imshape[0])]], dtype=np.int32)
cv2.fillPoly(mask, vertices, ignore_mask_color)
masked_image = cv2.bitwise_and(img, mask)
return masked_image
Functions of searching for lane lines either blindly or around the previously detected lanes¶
In [5]:
def find_lane_pixels_blind(binary_warped):
# HYPERPARAMETERS
# Choose the number of sliding windows
nwindows = 9
# Set the width of the windows +/- margin
margin = 100
# Set minimum number of pixels found to recenter window
minpix = 50
# Take a histogram of the bottom half of the image
histogram = np.sum(binary_warped[binary_warped.shape[0]//2:,:], axis=0)
# Create an output image to draw on and visualize the result
out_img = np.dstack((binary_warped, binary_warped, binary_warped))*255
# Find the peak of the left and right halves of the histogram
# These will be the starting point for the left and right lines
midpoint = np.int(histogram.shape[0]//2)
leftx_base = np.argmax(histogram[:midpoint])
rightx_base = np.argmax(histogram[midpoint:]) + midpoint
# Set height of windows - based on nwindows above and image shape
window_height = np.int(binary_warped.shape[0]//nwindows)
# Identify the x and y positions of all nonzero pixels in the image
nonzero = binary_warped.nonzero()
nonzeroy = np.array(nonzero[0])
nonzerox = np.array(nonzero[1])
# Current positions to be updated later for each window in nwindows
leftx_current = leftx_base
rightx_current = rightx_base
# Create empty lists to receive left and right lane pixel indices
left_lane_inds = []
right_lane_inds = []
# Step through the windows one by one
for window in range(nwindows):
# Identify window boundaries in x and y (and right and left)
win_y_low = binary_warped.shape[0] - (window+1)*window_height
win_y_high = binary_warped.shape[0] - window*window_height
win_xleft_low = leftx_current - margin
win_xleft_high = leftx_current + margin
win_xright_low = rightx_current - margin
win_xright_high = rightx_current + margin
# Draw the windows on the visualization image
#cv2.rectangle(out_img,(win_xleft_low,win_y_low),
#(win_xleft_high,win_y_high),(0,255,0), 5)
#cv2.rectangle(out_img,(win_xright_low,win_y_low),
#(win_xright_high,win_y_high),(0,255,0), 5)
# Identify the nonzero pixels in x and y within the window
good_left_inds = ((nonzeroy >= win_y_low) & (nonzeroy < win_y_high) &
(nonzerox >= win_xleft_low) & (nonzerox < win_xleft_high)).nonzero()[0]
good_right_inds = ((nonzeroy >= win_y_low) & (nonzeroy < win_y_high) &
(nonzerox >= win_xright_low) & (nonzerox < win_xright_high)).nonzero()[0]
# Append these indices to the lists
left_lane_inds.append(good_left_inds)
right_lane_inds.append(good_right_inds)
# If you found number of indices > minpix parameter, recenter next window on their mean position
if len(good_left_inds) > minpix:
leftx_current = np.int(np.mean(nonzerox[good_left_inds]))
if len(good_right_inds) > minpix:
rightx_current = np.int(np.mean(nonzerox[good_right_inds]))
# Concatenate the arrays of indices (previously was a list of lists of pixels)
try:
left_lane_inds = np.concatenate(left_lane_inds)
right_lane_inds = np.concatenate(right_lane_inds)
except ValueError:
# Avoids an error if the above is not implemented fully
pass
# Extract left and right line pixel positions
leftx = nonzerox[left_lane_inds]
lefty = nonzeroy[left_lane_inds]
rightx = nonzerox[right_lane_inds]
righty = nonzeroy[right_lane_inds]
# Fit a second order polynomial to each using `np.polyfit`
left_fit = np.polyfit(lefty, leftx, 2)
right_fit = np.polyfit(righty, rightx, 2)
return leftx, lefty, rightx, righty, out_img, left_fit, right_fit
In [6]:
def search_around_poly(binary_warped,left_fit, right_fit):
# HYPERPARAMETER
# Choose the width of the margin around the previous polynomial to search around it
margin = 100
# Grab activated pixels
nonzero = binary_warped.nonzero()
nonzeroy = np.array(nonzero[0])
nonzerox = np.array(nonzero[1])
# Set the area of search based on activated x-values
# within the +/- margin of our polynomial function
left_lane_inds = ((nonzerox > (left_fit[0]*(nonzeroy**2) + left_fit[1]*nonzeroy +
left_fit[2] - margin)) & (nonzerox < (left_fit[0]*(nonzeroy**2) +
left_fit[1]*nonzeroy + left_fit[2] + margin)))
right_lane_inds = ((nonzerox > (right_fit[0]*(nonzeroy**2) + right_fit[1]*nonzeroy +
right_fit[2] - margin)) & (nonzerox < (right_fit[0]*(nonzeroy**2) +
right_fit[1]*nonzeroy + right_fit[2] + margin)))
# Extract left and right line pixel positions
leftx = nonzerox[left_lane_inds]
lefty = nonzeroy[left_lane_inds]
rightx = nonzerox[right_lane_inds]
righty = nonzeroy[right_lane_inds]
# Fit a second order polynomial to each with np.polyfit()
left_fit = np.polyfit(lefty, leftx, 2)
right_fit = np.polyfit(righty, rightx, 2)
# Generate x and y values for plotting
ploty = np.linspace(0, binary_warped.shape[0]-1, binary_warped.shape[0])
# Calculate both polynomials using ploty, left_fit and right_fit
left_fitx = left_fit[0]*ploty**2 + left_fit[1]*ploty + left_fit[2]
right_fitx = right_fit[0]*ploty**2 + right_fit[1]*ploty + right_fit[2]
## Visualization ##
# Create an image to draw on and an image to show the selection window
out_img = np.dstack((binary_warped, binary_warped, binary_warped))*255
window_img = np.zeros_like(out_img)
# Color in left and right line pixels
out_img[nonzeroy[left_lane_inds], nonzerox[left_lane_inds]] = [255, 0, 0]
out_img[nonzeroy[right_lane_inds], nonzerox[right_lane_inds]] = [0, 0, 255]
# Generate a polygon to illustrate the search window area
# And recast the x and y points into usable format for cv2.fillPoly()
left_line_window1 = np.array([np.transpose(np.vstack([left_fitx-margin, ploty]))])
left_line_window2 = np.array([np.flipud(np.transpose(np.vstack([left_fitx+margin,
ploty])))])
left_line_pts = np.hstack((left_line_window1, left_line_window2))
right_line_window1 = np.array([np.transpose(np.vstack([right_fitx-margin, ploty]))])
right_line_window2 = np.array([np.flipud(np.transpose(np.vstack([right_fitx+margin,
ploty])))])
right_line_pts = np.hstack((right_line_window1, right_line_window2))
# Draw the lane onto the warped blank image
cv2.fillPoly(window_img, np.int_([left_line_pts]), (0,255, 0))
cv2.fillPoly(window_img, np.int_([right_line_pts]), (0,255, 0))
result = cv2.addWeighted(out_img, 1, window_img, 0.3, 0)
return leftx, lefty, rightx, righty, result, left_fit,right_fit
In [7]:
def fit_polynomial(binary_warped,is_First_Frame=True,left_fit=0,right_fit=0):
if (is_First_Frame == True):
# First frame so search blindly for lane lines
leftx, lefty, rightx, righty, out_img, left_fit, right_fit = find_lane_pixels_blind(binary_warped)
else:
# In case of video we already have history of where can we search for lane lines
leftx, lefty, rightx, righty, out_img, left_fit, right_fit = search_around_poly(binary_warped,left_fit,right_fit)
# Generate x and y values for plotting
ploty = np.linspace(0, binary_warped.shape[0]-1, binary_warped.shape[0] )
try:
left_fitx = left_fit[0]*ploty**2 + left_fit[1]*ploty + left_fit[2]
right_fitx = right_fit[0]*ploty**2 + right_fit[1]*ploty + right_fit[2]
except TypeError:
# Avoids an error if `left` and `right_fit` are still none or incorrect
print('The function failed to fit a line!')
left_fitx = 1*ploty**2 + 1*ploty
right_fitx = 1*ploty**2 + 1*ploty
## Visualization ##
# Colors in the left and right lane regions
out_img[lefty, leftx] = [255, 0, 0]
out_img[righty, rightx] = [0, 0, 255]
return out_img, left_fitx, right_fitx, ploty, left_fit, right_fit
In [8]:
def draw_lanes(image,warped, left_fitx, right_fitx, ploty, Minv):
# Create an image to draw the lines on
warp_zero = np.zeros_like(warped).astype(np.uint8)
color_warp = np.dstack((warp_zero, warp_zero, warp_zero))
# Recast the x and y points into usable format for cv2.fillPoly()
pts_left = np.array([np.transpose(np.vstack([left_fitx, ploty]))])
pts_right = np.array([np.flipud(np.transpose(np.vstack([right_fitx, ploty])))])
pts = np.hstack((pts_left, pts_right))
# Draw the lane onto the warped blank image
cv2.fillPoly(color_warp, np.int_([pts]), (0,255, 0))
# Warp the blank back to original image space using inverse perspective matrix (Minv)
newwarp = cv2.warpPerspective(color_warp, Minv, (image.shape[1], image.shape[0]))
return newwarp
Statistics Helper Functions¶
In [9]:
def measure_curvature_pixels(img_width, leftx, rightx, ploty):
'''
Calculates the curvature of polynomial functions in pixels.
'''
# Define conversions in x and y from pixels space to meters
ym_per_pix = 30/720 # meters per pixel in y dimension
xm_per_pix = 3.7/700 # meters per pixel in x dimension
# Define y-value where we want radius of curvature
# We'll choose the maximum y-value, corresponding to the bottom of the image
y_eval = np.max(ploty)
left_fit_cr = np.polyfit(ploty*ym_per_pix, leftx*xm_per_pix, 2)
right_fit_cr = np.polyfit(ploty*ym_per_pix, rightx*xm_per_pix, 2)
# Calculation of R_curve (radius of curvature)
left_curverad = ((1 + (2*left_fit_cr[0]*y_eval*ym_per_pix + left_fit_cr[1])**2)**1.5) / np.absolute(2*left_fit_cr[0])
right_curverad = ((1 + (2*right_fit_cr[0]*y_eval*ym_per_pix + right_fit_cr[1])**2)**1.5) / np.absolute(2*right_fit_cr[0])
curverad = int((left_curverad + right_curverad) / 2)
# Calculate left lane position
left_lane_pos = left_fit_cr[0]*(y_eval*ym_per_pix)**2 + left_fit_cr[1]*y_eval*ym_per_pix + left_fit_cr[2]
right_lane_pos = right_fit_cr[0]*(y_eval*ym_per_pix)**2 + right_fit_cr[1]*y_eval*ym_per_pix + right_fit_cr[2]
# Calculate middle point of 2 lanes
mid_lane_pos = (left_lane_pos+right_lane_pos)/2
img_center = int(img_width/2) * xm_per_pix
# Approximate to 2 digts, if car_pos this means car offset is to the right
# of the center, otherwise it is to the left of the center
car_pos = int((mid_lane_pos - img_center)*100+0.5) / 100
return curverad, car_pos
In [10]:
def print_statistics_to_image(final_image, curvature, car_offset):
# Build the curvature text with the curvature passed to the function
curvead_text = "Radius of Curvature : " + str(curvature) + "(m)"
# Build the car offset text with the offset passed to the function
if car_offset < 0:
offset_text = "Vehicle is " + str(car_offset * -1) + "m left of the center"
else:
offset_text = "Vehicle is " + str(car_offset) + "m right of the center"
# Add the text on top of the image
font = cv2.FONT_HERSHEY_DUPLEX
cv2.putText(final_image,curvead_text,(10,40), font, 1,(255,0,0),2,cv2.LINE_AA)
cv2.putText(final_image,offset_text,(10,75), font, 1,(255,0,0),2,cv2.LINE_AA)
Camera Calibration¶
1 - I'll compute the camera calibration using chessboard images¶
2 - I'll compute/save the camera matrix and distortion coefficients to be used later for distortion correction.¶
In [11]:
nx = 9
ny = 6
# prepare object points, like (0,0,0), (1,0,0), (2,0,0) ....,(6,5,0)
objp = np.zeros((6*9,3), np.float32)
objp[:,:2] = np.mgrid[0:9,0:6].T.reshape(-1,2)
# Arrays to store object points and image points from all the images.
# 3d points in real world space
objpoints = []
# 2d points in image plane.
imgpoints = []
# Make a list of calibration images
images = glob.glob('camera_cal/calibration*.jpg')
# Step through the list and search for chessboard corners
for fname in images:
img = cv2.imread(fname)
# Transfer to Gray Scale
gray = cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)
# Find the chessboard corners
ret, corners = cv2.findChessboardCorners(gray, (nx,ny), None)
# If found, add object points, image points
if ret == True:
objpoints.append(objp)
imgpoints.append(corners)
# Draw and display the corners
cv2.drawChessboardCorners(img, (nx, ny), corners, ret)
# Calibrate camera and calculate mtx and dist parameters
ret, mtx, dist, rvecs, tvecs = cv2.calibrateCamera(objpoints, imgpoints, gray.shape[::-1], None, None)
dst = cv2.undistort(img, mtx, dist, None, mtx)
# Draw Distorted vs Undistorted images
fig, axes = plt.subplots(nrows = 1,ncols=2, figsize=(20, 10))
axes[0].imshow(img)
axes[0].set_title('Distorted')
axes[1].imshow(dst)
axes[1].set_title('Undistorted')
# Save the camera calibration result for later use
dist_pickle = {}
dist_pickle["mtx"] = mtx
dist_pickle["dist"] = dist
pickle.dump( dist_pickle, open( "camera_cal/camera_cal_pickle.p", "wb" ) )
Image Pipeline consists of the following steps:¶
1 - Apply a distortion correction to raw images using camera calibration matrix and distortion coefficients previously calculated
2 - Calculate Adaptive threshold/Sobel threshold, H, L, S and generate combined image. I found adaptive threshold to be quite productive on the images.
3- Mark the region of interest from the combined image.
4 - Apply a perspective transform to rectify binary image ("birds-eye view").
5 - Detect lane pixels and fit to find the lane boundary.
6 - Determine the curvature of the lane and vehicle position with respect to center.
7 - Warp the detected lane boundaries back onto the original image.
8 - Output visual display of the lane boundaries and numerical estimation of lane curvature and vehicle position.
In [12]:
def Image_Pipeline(img):
bgr_img = cv2.imread(img)
# Convert to RGB Image
rgb_image = cv2.cvtColor(bgr_img, cv2.COLOR_BGR2RGB)
original_image = rgb_image
# Calculate Width and height of image
width = original_image.shape[1]
height = original_image.shape[0]
# Source Points for Image Warp
src_coordinates = np.float32(
[[280, 700], # Bottom left
[595, 460], # Top left
[725, 460], # Top right
[1125, 700]]) # Bottom right
dst_coordinates = np.float32(
[[250, 720], # Bottom left
[250, 0], # Top left
[1065, 0], # Top right
[1065, 720]]) # Bottom right
# Clear mtx and dist and load them from file camera_cal/camera_cal_pickle.p
dist = 0
mtx = 0
# load the camera matrix and distortion coefficient
dist_pickle = pickle.load( open( "camera_cal/camera_cal_pickle.p", "rb" ) )
mtx = dist_pickle["mtx"]
dist = dist_pickle["dist"]
# Undistort the image
undistort_image = cv2.undistort(original_image, mtx, dist, None, mtx)
# use cv2.getPerspectiveTransform() to get M, the transform matrix
M = cv2.getPerspectiveTransform(src_coordinates, dst_coordinates)
# Calculate Adaptive threshold, H, L, S and generate combined image
gray = cv2.cvtColor(undistort_image, cv2.COLOR_RGB2GRAY)
adaptive_Thresh = cv2.adaptiveThreshold(gray,255,cv2.ADAPTIVE_THRESH_GAUSSIAN_C, cv2.THRESH_BINARY,21,2)
hlsBinary_h = hls_threshold(undistort_image, channel='h', thresh=(5, 40))# HLS H-channel Threshold
hlsBinary_l = hls_threshold(undistort_image, channel='l', thresh=(200, 255))# HLS L-channel Threshold
hlsBinary_s = hls_threshold(undistort_image, channel='s', thresh=(90, 255))# HLS S-channel Threshold
# Combine channel thresholds
combined_image = np.zeros_like(hlsBinary_s)
combined_image[((hlsBinary_h == 1) & (hlsBinary_s == 1)) | (hlsBinary_l == 1) | (adaptive_Thresh == 1)] = 1
# Calculate Region of interest
roi_image = region_of_interest(combined_image)
# Use cv2.warpPerspective() to warp your image to a top-down view
warped_image = cv2.warpPerspective(roi_image, M, (width,height), flags=cv2.INTER_LINEAR)
out_img, left_fitx, right_fitx, ploty, left_fit, right_fit = fit_polynomial(warped_image)
Minv = cv2.getPerspectiveTransform(dst_coordinates, src_coordinates)
# Draw the lane lines found
newwarp = draw_lanes(original_image, warped_image,left_fitx,right_fitx, ploty,Minv)
final_image = cv2.addWeighted(undistort_image, 1, newwarp, 0.3, 0)
# Calculate lanes Curvature and car offset
curvead, pos = measure_curvature_pixels(final_image.shape[1], left_fitx, right_fitx, ploty)
# Print the statistics to the image
print_statistics_to_image(final_image, curvead, pos)
f, axes= plt.subplots(1,5,figsize=(15,30))
# Visualize Undistorted Image
axes[0].imshow(undistort_image)
axes[0].set_title("Undistorted Image")
# Visualize Warped Image
axes[1].imshow(combined_image)
axes[1].set_title("Combined Image")
# Visualize Region of Interest
axes[2].imshow(roi_image)
axes[2].set_title("ROI Image")
# Visualize Warped Image
axes[3].imshow(out_img)
axes[3].set_title("Warped Image")
# Visualize Final Image to be saved
axes[4].imshow(final_image)
axes[4].set_title("Final Image")
# Extract name of image to be stored from path
name = os.path.basename(img)
# Convert to BGR
bgr_img = cv2.cvtColor(final_image, cv2.COLOR_RGB2BGR)
# Save warped images to warped_images folder
cv2.imwrite('warped_images/' + name,out_img)
# Save final images to output_images folder
cv2.imwrite('output_images/' + name,bgr_img)
In [13]:
# Reading Images from test_images folder
test_images = glob.glob('test_images/*.jpg')
for fname in test_images:
Image_Pipeline(fname)
In [ ]:
In [15]:
def Video_Pipeline(img):
global is_First_Frame
global left_fit
global right_fit
original_image = img
# Calculate Width and height of image
width = original_image.shape[1]
height = original_image.shape[0]
# Source Points for Image Warp
src_coordinates = np.float32(
[[280, 700], # Bottom left
[595, 460], # Top left
[725, 460], # Top right
[1125, 700]]) # Bottom right
dst_coordinates = np.float32(
[[250, 720], # Bottom left
[250, 0], # Top left
[1065, 0], # Top right
[1065, 720]]) # Bottom right
# Clear mtx and dist and load them from file camera_cal/camera_cal_pickle.p
dist = 0
mtx = 0
# load the camera matrix and distortion coefficient
dist_pickle = pickle.load( open( "camera_cal/camera_cal_pickle.p", "rb" ) )
mtx = dist_pickle["mtx"]
dist = dist_pickle["dist"]
# Undistort the image
undistort_image = cv2.undistort(original_image, mtx, dist, None, mtx)
# Use cv2.getPerspectiveTransform() to get M, the transform matrix
M = cv2.getPerspectiveTransform(src_coordinates, dst_coordinates)
# Calculate Adaptive threshold, H, L, S and generate combined image
gray = cv2.cvtColor(undistort_image, cv2.COLOR_RGB2GRAY)
adaptive_Thresh = cv2.adaptiveThreshold(gray,255,cv2.ADAPTIVE_THRESH_GAUSSIAN_C, cv2.THRESH_BINARY,21,2)
hlsBinary_h = hls_threshold(undistort_image, channel='h', thresh=(5, 40))# HLS H-channel Threshold
hlsBinary_l = hls_threshold(undistort_image, channel='l', thresh=(200, 255))# HLS L-channel Threshold
hlsBinary_s = hls_threshold(undistort_image, channel='s', thresh=(90, 255))# HLS S-channel Threshold
# Combine channel thresholds
combined_image = np.zeros_like(hlsBinary_s)
combined_image[((hlsBinary_h == 1) & (hlsBinary_s == 1)) | (hlsBinary_l == 1) | (adaptive_Thresh == 1)] = 1
# Calculate Region of interest
roi_image = region_of_interest(combined_image)
# Use cv2.warpPerspective() to warp your image to a top-down view
warped_image = cv2.warpPerspective(roi_image, M, (width,height), flags=cv2.INTER_LINEAR)
# Fit a polynomial function ( Blind search(First frame) or Search around previously detected lanes)
out_img, left_fitx, right_fitx, ploty, left_fit, right_fit = fit_polynomial(warped_image,is_First_Frame, left_fit, right_fit)
Minv = cv2.getPerspectiveTransform(dst_coordinates, src_coordinates)
# Draw lane lines on the image
newwarp = draw_lanes(original_image, warped_image,left_fitx,right_fitx, ploty,Minv)
final_image = cv2.addWeighted(undistort_image, 1, newwarp, 0.3, 0)
# Calculate Curvature and Car offset from center
curvead, offset = measure_curvature_pixels(final_image.shape[1], left_fitx, right_fitx, ploty)
# Print statistics to the final image
print_statistics_to_image(final_image, curvead, offset)
# Last step clear the is_First_Frame flag we do not need to blindly search for lanes again,
# we can search around the previous detected lanes +/- a margin (hyper parameter)
is_First_Frame = False
return final_image
In [16]:
is_First_Frame = True
left_fit = 0
right_fit = 0
video_output1 = 'project_video_output.mp4'
video_input1 = VideoFileClip('project_video.mp4')#.subclip(0,10)
processed_video = video_input1.fl_image(Video_Pipeline)
%time processed_video.write_videofile(video_output1, audio=False)
In [ ]: